A recurring theme in Yale OAIC studies of multifactorial geriatric health conditions is the determining the mechanism(s) of action of multicomponent interventions. This is a complicated analytic problem in which little methodology has been developed to date; thus, warranting a targeted development project. To investigate this analytic problem, we will use data from two recently completed Yale OAIC interventional development studies, DRIVER and PREHAB, which both used a physical therapy program. Even though DRIVER and PREHAB both had a multi-component physical ability intervention, the interventions differed in that DRIVER's was global (i.e., all participants received the full intervention) and PREHAB's was individualized (i.e., participants received an individualized intervention that focused only on the impairments found to be deficient at baseline). The respective study populations and outcomes also differed. The PREHAB population was frailer than DRIVER's population and the outcome for DRIVER was a measure of driving performance while PREHAB used a measure of frailty. In both studies it was hypothesized that the physical conditioning program would improve physical ability and result in improved outcomes. Thus, these two studies provide the opportunity to develop the methodology in one study (DRIVER) and test it in the other (PREHAB). If the methodology for DRIVER is also applicable to PREHAB, it is likely to be generalizable to other types of studies because these two studies have differences in their respective designs. In contrast, if the methodology developed for DRIVER is not applicable to PREHAB, the differences in the respective study designs will provide an opportunity to determine the reasons why the methodology is not generalizable and how to possibly refine the methods. Thus, these two studies provide a unique opportunity to develop and test methodology for investigating mechanisms of action and determine whether these results can be translated into the design of more efficient OAIC studies.